ED2LM: Encoder-Decoder to Language Model for Faster Document Re-ranking Inference.
Kai HuiHonglei ZhuangTao ChenZhen QinJing LuDara BahriJi MaJai Prakash GuptaCícero Nogueira dos SantosYi TayDon MetzlerPublished in: CoRR (2022)
Keyphrases
- word clouds
- language model
- document ranking
- ad hoc information retrieval
- language modeling
- video codec
- document length
- distributed video coding
- smoothing methods
- document retrieval
- n gram
- information retrieval
- probabilistic model
- query expansion
- relevance model
- ranking functions
- pseudo feedback
- query specific
- language modelling
- retrieval model
- test collection
- mixture model
- human relevance judgments
- vector space model
- document representation
- inter document similarities
- document level
- query terms
- expert search
- pseudo relevance feedback
- dirichlet prior
- bayesian networks
- feature selection
- context sensitive
- term dependencies
- relevant documents
- retrieval systems
- web documents
- document collections
- search engine